To take it one step further, I don't necessarily trust studies linking genomic variants with disease significance. A subset derive correlations using oversimpli

To take it one step further, I don't necessarily trust studies linking genomic variants with disease significance. A subset derive correlations using oversimplified models: http://www.ncbi.nlm.nih.gov/pubmed/23770567 http://www.ncbi.nlm.nih.gov/pubmed/23770567